|
Description |
|
Cloud computing and virtualization are currently hot topics and have been generating substantial interest in the
community and it is anticipated that this interest will expand. It is also important to the management community
as it is challenging traditional management methods by its sheer size and high level of automation. Clouds must
provide appropriate levels of performance to large groups of diverse users, and those clouds are accessed
through virtualized wide area networks. Management systems are essential for that and thereby for the future
success of the cloud paradigm. New systems, methods, and approaches for cloud and virtualization
management from an IT as well as a networking perspective are to be discussed at this workshop.
|
|
|
Program |
|
| 08:45 - 10:30 - Session 1 - Chair: Bruno Schulze |
Obfuscatory obscanturism: making workload traces of commercially-sensitive systems safe to release
Charles Reiss, John Wilkes, Joseph Hellerstein |
| Abstract: Cloud providers such as Google face daunting technical challenges in supporting planetary-scale distributed systems. Although they are interested in fostering research on these challenges, no academic organizations have similar scale systems on which to experiment. Fortunately, good research can still be done using traces of real-life production workloads. However, there are considerable risks in releasing such data. For example, it might inadvertently disclose confidential or proprietary information, such as happened with the Netflix Prize data. This paper discusses our approach to this problem, systematic obfuscation, and our experiences using it to publish a month-long trace of a production system workload. Systematic obfuscation protects proprietary and personal data while leaving it possible to answer interesting research questions. We discuss several of the design issues that this approach deals with, and explain some of the risks, and how they can be mitigated. |
Workload Characterization and Prediction in the Cloud: A Multiple Time Series Approach
Arijit Khan, Xifeng Yan, Shu Tao, Nikos Anerousis |
| Abstract: Cloud computing promises high scalability, flexibility and cost-effectiveness to satisfy emerging computing requirements. To efficiently provision computing resources in the cloud, system administrators need the capabilities of characterizing and predicting workload on the Virtual Machines (VMs). In this paper, we use data traces obtained from a real data center to develop such capabilities. First, we search for repeatable workload patterns by exploring cross-VM workload correlations resulted from the dependencies among applications running on different VMs. Treating workload data samples as time series, we develop a co-clustering technique to identify groups of VMs that frequently exhibit correlated workload patterns, and also the time periods in which these VM groups are active. Then, we introduce a method based on Hidden Markov Modeling (HMM) to characterize the temporal correlations in the discovered VM clusters and to predict variations of workload patterns. The experimental results show that our method can not only help better understand group-level workload characteristics, but also make more accurate predictions on workload changes in a cloud. |
Developing and Managing Customizable Software as a Service Using Feature Model Conversion
Hendrik Moens, Eddy Truyen, Stefan Walraven, Wouter Joosen, Bart Dhoedt, Filip De Turck |
| Abstract: In recent years, there has been a growing interest in cloud technologies. Using current cloud solutions, it is however difficult to create customizable multi-tenant applications, especially if the application must support varying Quality of Service (QoS) guarantees. Software Product Line Engineering (SPLE) and feature modeling techniques are commonly used to address these issues in traditional desktop applications, but these techniques cannot be ported directly to a cloud context as such changes are usually compiled into the application. In this paper, we propose an architecture for the development and management of customizable Software as a Service (SaaS) applications, built using SPLE techniques. In our approach, each application is composed as a Service-Oriented Architecture (SOA) in which individual services correspond to application functionalities, referred to as features. A feature-based methodology is described to abstract and convert the application information required at different stages of the application life-cycle: development, customization and execution. |
Monitoring Abnormal Behaviors for Mobile Cloud Services
Taehyun Kim, Yeongrak Choi, Jae Yoon Chung, Jian Li, Jonghwan Hyun, James Hong |
| Abstract: Recently, various mobile cloud services are increased rapidly, and they are being expanded to mobile area. We focus on a convergence service between mobile and cloud services, which is based on the virtualization of mobile devices. Security is the most important issue for mobile cloud service and should be solved first to commercialize it. In this paper, we define mobile cloud service and discuss service scenarios to be aware of vulnerabilities ranges from mobile devices to cloud infrastructure. We also propose a methodology to monitor user behaviors and detect abnormal usages. |
| 11:00 - 12:45 - Session 2 - Chair: Omar Cherkaoui |
Environment for Automatic Genrations of Platforms in Cloud Computing
Hélder Borges, Bruno Schulze, José De Souza, Antonio Mury |
| Abstract: Cloud computing has been established in recent years as an important platform for research. In the current scenario, tasks such as obtaining, sharing, manipulation and exploitation of large amounts of data are common and they require many resources. Cloud computing can contribute to this scenario because it can provide indefinitely processing, memory, storage, and others resources for immediate use. Considering this context, we have that ponder on the infrastructure of the cloud environment, and for this purpose, the use of models can help because the models may contain information about the user, hardware and software that could be used by computational mechanisms to automatically build all the infrastructure necessary for the operation of an virtual machine in a cloud computing environment. Moreover, the models can be exported to other formats used by cloud services providers. Service Level Agreements - SLA can be used to control the utilization of computational resources from a provider in cloud computing environments and still guarantee the quality of services. The purpose of this work is to combine these paradigms and technologies, with the purpose of creating an environment for automatic generation of platforms in cloud computing. |
Kagemusha: A Guest-Transparent Mobile IPv6 Mechanism for Wide-Area Live VM Migration
Takahiro Hirofuchi, Hidemoto Nakada, Satoshi Itoh, Satoshi Sekiguchi |
| Abstract: We are developing a wide-area live migration mechanism that allows dynamic load balancing of virtual machines (VMs) among datacenters. We consider that Mobile IPv6 (MIPv6) is a promising technology to support transparent network reachability when VMs migrate to foreign networks. Existing MIPv6 mecha- nisms, however, are not suitable for VM migrations; real-world IaaS datacenters require guest-transparent and exible tunneling mechanisms, which are not provided by existing MIPv6 programs. In this paper, we propose a guest-transparent MIPv6 tunneling mechanism (Kagemusha), which performs Client MIPv6 signaling and tunneling on a host operating system. No MIPv6 program is required to be installed into a guest operating system. The proposed system is fully compatible with existing home agents (HAs). It basically works with any virtual machine monitors. Through experiments, we confirmed that our prototype system successfully established MIPv6 tunnels with HAs, and its performance overhead was negligible for normal use cases. We also confirrmed that the prototype system successfully worked with live migrations; the downtime of migration increased only by several hundred milliseconds. |
Autonomic Cloud Resource Scaling
Edgar Magana, Masum Hasan, Alexander Clemm, Sree Gudreddi |
| Abstract: A Cloud is a very dynamic environment where resources offered by a Cloud service provider, out of one or more Cloud DCs are acquired or released by an enterprise on-demand and at any scale. Given the on-demand nature and the scale at which a Cloud operates, there is need for automation for all the Cloud management functions (provisioning, orchestration, monitoring, resource scaling, etc.) of a Cloud. In a Cloud, resources from three separate domains, compute, storage and network, are acquired or released on-demand. Hence any management function should integrate aspects of these three domains for effective management and operations of a Cloud. Typically this is not the case, especially for Cloud resource scaling, where resources from a particular domain are scaled individually, but not in relation to other domains. For example, while scaling compute resources, network related metrics are not correlated with compute metrics in making a scaling decision. In this paper we describe a framework that integrates metrics from multiple domains in making scaling decisions. In addition, it also scales network resources together with resources from the other domains. The framework also supports policy-based automation of scaling, where policies are based on policies related to the all three domains. |
|
|
Deadlines
|
|
Paper registration due: December 14, 2011
Paper submission: December 20, 2011
Notification of acceptance: January 31, 2012
Final camera-ready papers due: January 20, 2012
Final version of papers due: February 15, 2012
Workshop dates: April 20, 2012 |
|
Topics
|
|
For the CloudMan 2012 workshop researchers from the Cloud Management and Virtualization communities are
encouraged to submit and present original work to be considered for publication. Overall topics of interest
include but are not limited to:
- Cloud computing environments
- Cloud service orchestration
- Cloud APIs and usage control
- Cloud data management
- Cloud scalable monitoring
- Cloud load balancing
- Cloud federation management
- Customer cloud management
- Managing data centers
- Management as a service
- Management of virtual slices
- Optimization of data center and workload energy consumption
- IaaS, Paas, Saas, NaaS management
|
- Accounting and economic models for clouds
- Managing cloud services
- Management of virtualized hardware resources
- Network-specific mechanisms for optimized cloud access
- Performance modeling & evaluation
- QoS/QoE management in the cloud
- Management tools for infrastructure virtualization
- Automated resource slicing
- Integration of the wireless and optical domains
- Policy driven service/resource life-cycle management
- Applications and services enabled by virtualized infrastructure
|
|
|
Paper Submission
|
|
The Cloud Management workshop invites authors to submit original and unpublished
work. Papers should not exceed 8 pages IEEE style (single-spaced 2-column text using 10-point size type on
A4 paper). Authors should submit a PostScript (level 2) or PDF file that will print on a PostScript printer.
- Electronic submission only (via submission server)
- Submission requires the willingness of at least one of the authors to register and present the paper.
- All selected papers for this workshop are peer-reviewed and will be published with IEEE Xplore.
|
|
|
Organization |
Workshop Co-Chairs:
Bruno Schulze – LNCC, BR
Peer Hasselmeyer – NECLAB, DE
Omar Cherkaoui – UQAM, CA |
Publicity Co-Chairs:
Carlos Westphall – UFSC, BR
Laurent Mathy – London College, UK
Marcus Brunner – NEC, DE
Sergi Figuerola – i2cat, ES |
|
|
Program Committee (Preliminary) |
Akihiro Nakao – University of Tokyo, JP
Antonio Mury – LNCC, BR
Antonio Tadeu Gomes – LNCC, BR
Dae Young Kim – Chungnam National Univ., KR
Dipankar Raychaudhuri – Rutgers Univ., US
Dominique Dudkowski – NEC, DE
Edmundo Madeira – UNICAMP, BR
Fabio Porto – LNCC, BR
Guy Pujolle – Lip6 / ParisTech, FR
Hassan Massum – Cisco, US
Jennifer Rexford – Princeton Univ., US
Jose Neuman – UFC, BR
Leo Garcia – Univ. of Toronto, CA
Lisandro Granville - UFRGS, BR
Luis Carlos Erpen de Bona – UFPR, BR
|
Mathieu Lemay – Inocybe, CA
Michael Bauer – Univ. of Waterloo, CA
Nick McKeown – Stanford Univ., US
Noemie Simoni – ENST Paristech, FR
Odej Kao – TU Berlin, DE
Olivier Festor – INRIA Grand Nancy, FR
Pascale Primet – INRIA, Univ. of Lyon, FR
Proposer Chemoul – France Telecom, FR
Raouf Boutaba – Univ. of Waterloo, CA
Stephan Baucke – Ericsson, DE
Sue Moon – KAIST, KR
Thomas Magedanz – FOKUS/FHG, DE
Ulas Kozat – Docomo Labs, US
Wessam Ajib – UQAM, CA
Yves Lemieux – Ericsson, CA
|
|
|